#codex

RSS Feed
AI Hacker News Apr 6, 2026 2 min read

OpenAI's current Codex rate card now maps credit usage to input, cached-input, and output tokens instead of relying only on rough per-message estimates. The April 5, 2026 Hacker News thread focused on how that gives teams clearer cost visibility, while also leaving a split world where some plans use the new token-based card and others remain on the legacy message-based card until migration.

LLM sources.twitter Mar 29, 2026 2 min read

OpenAIDevs pointed developers to Codex Security on March 29, 2026, positioning it as a way to find, validate, and remediate likely vulnerabilities in connected GitHub repositories. OpenAI's docs say the system scans commit by commit, uses repo-specific threat models, validates high-signal findings in an isolated environment, and can move reviewed findings toward GitHub pull requests.

© 2026 Insights. All rights reserved.